A comparison of methods for evaluating soil redistribution in the severely eroded Stavropol region, southern European Russia

In this paper, we combined several erosion assessment methods to construct a sediment budget describing soil redistribution and sediment delivery within a study area containing grassed upper slopes, a large arable field of 1.3 km2 with a semipermanent rill and ephemeral gully network and a downslope buffer zone of a grassed dry valley (balka) bank with depositional fans. The study site is in the Stavropol Upland-one of the most severely eroded, intensively cultivated areas of European Russia. The methods include two variations of the soil survey approach; a proportional 137Cs conversion model; a mass balance 137Cs conversion model; a 137Cs-based tracer budget; direct measurement of gully volume by theodolite; examination of 137Cs depth profiles; and a version of the USLE model modified and calibrated for Russian conditions. Our results highlight the importance of (i) comparing such techniques, (ii) validating the results from them, and (iii) the value of combining the outputs of different measurement methods. In particular, the soil survey approach was able to separate the influence of sheet and linear erosion; the proportional 137Cs and mass balance 137Cs models estimated similar soil redistribution rates (5.5±0.8 and 5.3±0.8 kg m−2 year−1, respectively) and were improved when combined with direct measurements of gully volumes. Rates and locations of sediment redeposition within sinks, such as grassed valley banks, were best evaluated by combining 137Cs depth profile analysis and conversion models with soil profile descriptions. There was good agreement between the soil survey and the 137Cs tracing (combined with gully volume measurement) approaches. Average erosion rates estimated using the Russian version of the USLE model were lower by a factor of six compared to the physically based approaches. It may have been successful in assessing water erosion rates within inter-rill areas, and the discrepancy may provide insight into the contribution of tillage erosion. We conclude that the USLE model should be used only in combination with other techniques on arable fields where intensive rill erosion, ephemeral gullying, and mechanical translocation of soil take place.